Objective: To quantify regional and unit variations in antibiotic use during the first week of life, incidence of early-onset sepsis (EOS) and mortality in late-preterm and term newborns in Sweden.
Design: A nationwide cohort study.
Setting: The Swedish Neonatal Quality Register, a register of all newborns admitted for neonatal care in Sweden.
Patients: All late-preterm and term newborns (≥34 weeks' gestation) admitted to a neonatal unit from 1 January 2012 to 31 December 2020.
Main outcome measures: The proportion of antibiotic use, incidence of EOS and mortality in late-preterm and term newborns.
Results: Out of 1 025 515 newborns, 19 286 neonates (1.9%) received antibiotics with a 2-fold to 5-fold variation across regions (1.3%-3.0%) and units (0.9%-4.3%). Duration of antibiotic therapy (median) varied across regions from 7-10 days for infants with EOS, and 4-7 days for infants with no sepsis. Incidence of EOS ranged from 0.33 to 0.93 per 1000 live births between regions. The number of infants treated per EOS case varied 2-fold to 15-fold across regions (22-39) and units (7-113). All-cause mortality and EOS-associated mortality ranged across regions from 0.21 to 0.54 per 1000 live births and 0.0 to 34.9 (3 of 86 newborns) per 1000 infants with EOS.
Conclusions: This nationwide study revealed wide variations in antibiotic use and in the number of infants treated per EOS case. The results indicate a disproportionate antibiotic use in relation to the incidence of EOS. This emphasises the need for future efforts to minimise unwarranted antibiotic use.
Keywords: Epidemiology; Intensive Care Units, Neonatal; Neonatology; Sepsis; Therapeutics.
© Author(s) (or their employer(s)) 2026. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ Group.